Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces a novel approach for vocal intensity conversion that allows for continuous control of target sound pressure level (SPL) on a dB scale, moving beyond discrete categories or reference utterances. Utilizing a conditional cycle-consistent GAN trained on non-parallel data, the system converts features from a frozen speech encoder, which are then resynthesized by a pretrained vocoder. Experiments with WavLM features demonstrate significant achieved intensity change and fundamental frequency shifts, validated by both objective measurement and perceptual evaluation.
Continuous vocal intensity conversion is now achievable on a dB scale, enabling precise control over speech loudness without parallel data.
Vocal intensity conversion renders a recorded utterance at a different vocal intensity, rather than simply scaling its amplitude, while preserving its content and speaker identity. Existing systems control intensity through a reference utterance or a few discrete categories. Instead, we control the conversion with any target sound pressure level (SPL) on a continuous dB scale. A lightweight converter, trained as a conditional cycle-consistent GAN without parallel data, transforms the features of a frozen speech encoder, from which a pretrained vocoder resynthesizes the speech. We compare three representations, log-mel spectrograms, WavLM, and SpeechTokenizer, and introduce an objective protocol that measures the achieved intensity change against the target one on a calibrated corpus. With WavLM, the converter achieves a measured 83% of the target change in intensity and also shifts the fundamental frequency. A perceptual evaluation confirms that perceived vocal intensity increases with the target level on amplitude-normalized stimuli.